Children with Disabilities in Poor Households: Association with Juvenile and Adult Offending

Background: Among youth transitioning from special education, 19% were arrested 2 years following school (Doren, Bullis, & Benz, 1996). For those labeled emotionally disturbed, 58% are arrested at least once (Wagner, Newman, et al, 2006). It is unknown how this varies by poverty status. About 16% of low-income families have a disabled child (Lee, Sills & Oh, 2002) and poor children in special education have higher rates of behavioral problems while in school (OSEP, 2003). Little is known about how poverty and disability are associated with behaviors outside school and into adulthood. Further, existing studies do not control for other risks like maltreatment.

Method: Using data from a larger longitudinal study of poor compared to maltreated children (see Jonson-Reid et. al, 2005), this study helps fill this gap in knowledge. Children (born 1982 through 1986) who received AFDC/TANF at least once, but have no record of maltreatment are followed through mid-year 2006 (N=2158). Cox regression adjusted for clustering by census tract was used to examine the link between disability (measured by eligibility for special education- 20.3% of sample) and delinquency or later adult crime among poor children without known maltreatment. Due to small sample size for certain disability types, disability was categorized as: (1) learning disabled, (2) emotional disturbance and other health impaired; (3) other disability; (4) no known disability. Dichotomous dependent variables included “juvenile delinquency petition (18.72% for children with disabilities vs. 12.79% for children without)”; “adult arrest (13.93% vs. 9.24%)”; and “entry into adult corrections (9.59% vs. 5.81%)”. Using an ecological framework analyses controlled for (1) child age, gender, and race; (2) parent age, education, employment status, (3) high school graduation rate in census tract; and (4) service variables: parent mental health service record, welfare program participation (duration of welfare participation and re-entry into welfare programs), and delinquency (in models of adult outcomes).

Results: Children labeled as learning disabled had higher rates of juvenile offending. Children with labels of emotional disability or other health impaired had higher risk of adult corrections (hazard ratio=5.73, p<.001) and adult arrest (hazard ratio=2.00, p<.01). Longer duration on welfare and being male were associated with higher risk of juvenile and adult offending. Juvenile offending was associated with over two times higher risk of adult offending regardless of disability.

Implications: Rates of offending behavior for children with disabilities were higher than similarly poor peers. Research indicates that children with disabilities who need mental health care often do not receive it (Witt, Kasper & Riley, 2003). National data indicate less than 53% of schools offer such services to children with disabilities (OSEP, 2003). The juvenile court also seems a likely target for intervention, yet this system also lacks adequate mental health services (Leone, 1994) and few first time juvenile offenders receive any formal intervention (OJJDP, 2003). Our findings support increasing the availability of social work services in schools and the juvenile justice system to help offset offending behavior for poor children with disabilities.